Cron Scheduler Pro turns natural-language scheduling intentions into reliable, auditable, and locally stored task execution contracts for AI Agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cron-scheduler-pro
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install cron-scheduler-pro using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Cron Scheduler Pro?
Cron Scheduler Pro is a local-first scheduling engine for AI Agents that converts requests such as “run this every two hours” into dependable daily, weekly, monthly, interval, or one-time jobs. It is designed for Openclaw Skills workflows that need predictable timing without cloud synchronization or third-party cron services.
The skill addresses timezone drift, missed tasks, accumulated one-time jobs, concurrent writes, and silent failures. It provides timezone validation, next-run previews, atomic file updates, automatic cleanup, execution history, retry backoff, circuit breakers, and task lifecycle controls while relying on Python 3.8+ and standard libraries.
Cron Scheduler Pro Use Cases
- Schedule recurring inbox checks, data synchronization, maintenance scripts, or Agent-generated summaries.
- Generate daily reports, briefings, health checks, and operational dashboards at a fixed local time.
- Create precise one-time reminders that automatically archive after successful execution.
- Replace expensive heartbeat polling with lightweight due-task checks using
next_run.py --due-only. - Run API or endpoint health checks with bounded retries and automatic circuit-breaker pausing.
- Support teams across time zones by assigning an explicit timezone to each job.
- Coordinate local automation for personal workflows, solo businesses, and Agent orchestration pipelines.
- Use Openclaw Skills when local persistence, auditability, and predictable scheduling matter more than multi-device synchronization.
How Cron Scheduler Pro Works
- Confirm the runtime environment and lock the default timezone before creating the first job.
- Parse a natural-language request or structured parameters into a job definition with a name, schedule type, execution time, timezone, and task payload.
- Validate the timezone and schedule configuration; daily, weekly, and monthly jobs require clear time semantics.
- Write the complete job atomically under an exclusive file lock, avoiding unsafe add-then-update operations.
- Calculate and expose the next scheduled run so an Agent or user can preview when the task will execute.
- Trigger active jobs when the Agent heartbeat or scheduler checks for due work.
- Record start time, finish time, duration, status, retries, and errors in the local run history.
- Retry failures using the configured backoff schedule; after the failure threshold is reached, pause the job and notify the user.
- Move successful one-time jobs from
activetoarchived, retain their run history, and physically remove old archives during cleanup. - Manage jobs through the
active,paused, andarchivedlifecycle, with statistics and troubleshooting data available locally.
Cron Scheduler Pro Setup
Requirements
- An AI Agent platform that supports
SKILL.mdinstructions and can execute Python scripts. - Python 3.8 or newer.
- Linux, macOS, or Windows. Linux and macOS use
fcntl; Windows uses the standard-librarymsvcrtequivalent. - No external API key is required by the scheduling engine. A scheduled task may independently require credentials for the service it calls.
Installation and timezone configuration
Place the skill and its tools/cron/ scripts in the Agent workspace, verify Python, then configure the timezone before creating jobs:
python3 --version
python3 tools/cron/set_timezone.py --timezone "Asia/Shanghai"
The timezone should also be recorded in MEMORY.md:
Timezone: Asia/Shanghai (UTC+8)
Create a recurring job
python3 tools/cron/add_job.py \
--name "daily-standup" \
--type daily \
--time "09:00" \
--timezone "Asia/Tokyo" \
--task "Send the daily standup reminder"
Supported schedule types include daily, weekly, monthly, interval, and once. For interval jobs, define the cadence with a value such as 2h or 5m:
python3 tools/cron/add_job.py \
--name "inbox-check" \
--type interval \
--every "2h" \
--task "Check the inbox and provide a summary"
Preview and inspect jobs
python3 tools/cron/next_run.py --due-only
python3 tools/cron/stats.py
Clean up archived jobs
One-time jobs are automatically archived after successful execution. To remove archives older than seven days and optionally create a backup, use:
python3 tools/cron/cleanup.py --older-than 7d
python3 tools/cron/cleanup.py --backup
Use the platform’s pause_job.py or archive_job.py tools to cancel or deactivate a job before it runs. The scheduler depends on the Agent heartbeat to wake and execute due tasks; it is not an independent system daemon.
Cron Scheduler Pro Data Schema & Taxonomy
Local storage
All scheduler data is stored locally under:
$HOME/.skill-platform/workspace/memory/cron/
├── jobs.json # Job definitions and lifecycle state
├── runs.json # The most recent 100 execution records
├── stats.json # Cached success, failure, and duration statistics
└── .lock # Exclusive file lock for concurrent writes
The timezone is persisted separately in MEMORY.md. There is no built-in cloud sync, third-party cron service, or cross-device replication.
Job lifecycle
| State | Meaning | Triggered? |
|---|---|---|
active |
Scheduling is enabled | Yes |
paused |
Temporarily disabled or automatically paused by the circuit breaker | No |
archived |
No longer active but retained for history | No |
Schedule taxonomy
| Type | Example | Behavior |
|---|---|---|
daily |
Every day at 09:00 | Fixed daily local time |
weekly |
Monday at 10:00 | Fixed weekly time |
monthly |
The first day at 08:00 | Fixed monthly time |
interval |
Every 2 hours | Relative to creation time unless aligned |
once |
2026-07-18T15:30:00+08:00 |
Runs once and archives after success |
Execution record
Each run contains fields such as job_name, started_at, finished_at, duration_ms, status, retries, and error. Run history is capped at 100 records, while cached statistics summarize recent success rate, failure count, and average duration.
One-time job metadata
One-time jobs enforce delete_after_run: true. After a successful run, the job moves to archived, remains available for seven days, and is then eligible for physical cleanup. The execution record remains in runs.json until history retention removes it.
Cron Scheduler Pro Advanced Features
- Mandatory timezone locking and per-job timezone overrides reduce ambiguous “9 AM” schedules and daylight-saving surprises.
- Atomic temporary-file writes followed by rename protect
jobs.jsonfrom partial updates. - Exclusive file locking prevents concurrent Agent processes from corrupting local scheduler state.
- Retry policies support up to three retries with configurable backoff, such as 60, 300, and 900 seconds.
- Circuit breakers automatically pause jobs after five consecutive failures and record the reason for later notification.
- Next-run previews support batch calculation and lightweight
--due-onlyheartbeat checks. - One-time jobs self-archive and receive automatic retention cleanup to prevent schedule accumulation.
- Pause, resume, archive, and cancellation workflows support operational control without deleting execution history.
- Cached statistics expose recent run volume, success percentage, failures, average duration, and error details.
- Per-user local storage provides isolation by default, while shared directories can be used with appropriate local filesystem locking.
- The design works with multiple AI Agents as independent heartbeat-driven workers, although shared scheduling requires a shared local directory.
- Data can be migrated to another machine by copying the
cron/directory together with the timezone entry inMEMORY.md. - The engine is compatible with Openclaw Skills patterns for local automation, but it is not intended for strict, safety-critical decisions requiring absolute determinism.
- Network filesystems such as NFS or SMB are not recommended because their file-lock behavior may be unreliable.
SKILL.md
Loading
Related Openclaw Skills
Ticket Revenue Settlement Check (Free)
A free local checker that recomputes ticket revenue, channel commissions, refunds, and settlement totals line by line.
Grocery Price Book
Grocery Price Book turns receipts into personalized price intelligence for comparing stores, detecting shrinkflation, and making smarter bulk-buy decisions.
Chemical Inventory Check (Free)
A deterministic, offline CLI checker that reconciles hazardous-chemical inventory, requisition, return, and counting ledger data across six core validation rules.
Subcontractor Qualification Check (Free)
A free, deterministic offline validator that checks subcontractor qualification and safety-permit registers across six evidence-based rules.
Business Automation Hub
Business Automation Hub helps small teams decide what to automate, build reliable workflows, quantify ROI, and detect silent failures.
NeuroCache Pro
NeuroCache Pro gives AI agents graph-based associative memory for discovering relationships, tracing causes, resolving contradictions, and reusing knowledge across projects.
Featured*
